Horizon Europe Work Packages, Gantt and PERT: A Complete Worked Example

Every guide to Horizon Europe work plans explains the rules. Very few show a work plan in which all the rules have been applied at once, so that the tables agree with each other, the person-months add up and the chart matches the deliverable list. This page is that example. SALTMARSH is invented, but every column follows the current template, every number reconciles, and the same project is costed line by line in the companion page on building the lump-sum budget table.

What section 3.1 asks for

Section 3.1 of the Part B template, Work plan and resources, has an indicative 12 pages, or 17 for lump-sum topics, and those pages include the tables. The template asks for a brief presentation of the overall structure of the work plan, the timing of the work packages and their components as a Gantt chart or similar, a graphical presentation of how the components inter-relate as a PERT chart or similar, and eight tables. It also advises a distinct project-management work package and visible tasks or work packages for data management, dissemination and exploitation, and communication. For the full walk through the rest of the template, see the Part B template, section by section.

The template defines the three objects the tables describe. A deliverable is a report or other output sent to the Commission or agency to ensure effective monitoring. A milestone is a control point that helps chart progress, may correspond to a key result that lets the next phase begin, and must be verifiable. A critical risk is a plausible event that could have a high adverse impact on the project’s ability to achieve its objectives. The glossary has the other terms used below.

The example project: SALTMARSH

SALTMARSH, Scalable Adaptive Living Tools for Marsh And Riverine Shoreline Habitats, is a 42-month Research and Innovation Action under Cluster 6, funded as a Type 2 lump sum. It designs nature-based living shorelines, tests them in a hydrodynamic flume, installs them at three pilot sites in Ravenna, Aveiro and Lesvos, and turns two years of monitoring into evidence coastal authorities can use to procure the designs. The consortium has eight partners from six countries, three of them widening countries, and the reporting periods end at months 18, 36 and 42.

The project has six work packages. That sits inside the usual six-to-eight range, keeps each work package a substantial part of the plan, and follows the customary order of management first and dissemination last, with the scientific work packages in the sequence the work actually flows.

Table 3.1a: list of work packages

Two checks apply to this table before anything else. The person-months per work package must equal the column totals of table 3.1f, and project management should be a modest share of total effort: WP1 here is 34 of 424 person-months, or 8 percent.

Table 3.1b: one work package described in full

The template asks, for each work package, for its objectives and a description of work broken into tasks, naming the lead partner and the role of the other participants, with each task quantified so the resources can be justified and progress monitored. Deliverables are not repeated here because they are in table 3.1c. WP4 is the largest work package and the one with the most partners, so it is shown in full.

The per-task effort in the box adds up to the WP4 column of table 3.1f for every partner: AegCRI 14, ULV 10, DELTA 18, ILV 36, BWT 6, NORS 8, RAV 14 and AEA 12, 118 person-months in total. That reconciliation is the single most common thing evaluators find broken. The other five work packages follow the same pattern; their tasks are listed below in short form.

Table 3.1c: deliverables

Only deliverables essential for monitoring belong here, because each becomes a contractual obligation. SALTMARSH has fourteen, two or three per work package, numbered by work package and listed in order of delivery date. The data management plan and the plan for dissemination, exploitation and communication are both due by month 6, as the template requires.

Table 3.1d: milestones

A three-year project usually carries around five milestones; SALTMARSH runs 42 months and has six. Each one is a state of the project that is true or false on a given month, and each names the artefact that proves it. A useful sentence pattern is MSx: what is true, by month n, verified by which document. MS3 and MS4 are also the gates on the critical path in the PERT chart below.

Table 3.1e: critical risks

These are risks to implementation, not the external barriers to impact that belong in section 2.1. The template asks for likelihood and severity on a low, medium, high scale, with likelihood assessed after the mitigation is in place. Every mitigation below points to a task that already has effort assigned, which is what makes it credible; a mitigation with no budget behind it reads as a promise. For a structured way to score and rank them, use the risk assessment matrix.

Table 3.1f: summary of staff effort

The staff-effort matrix is where evaluators check that every partner has a real role and that effort matches ambition. The work package leader’s figure is shown in bold, as the template asks.

The Gantt chart

The Gantt shows each work package as a pale bar and its tasks as darker bars beneath, with deliverables as diamonds and milestones as triangles on the work package row. The dashed lines at months 18 and 36 mark the ends of the first two reporting periods. In a lump-sum project those lines matter for cash flow, not only for reporting, because a work package is paid only at the end of the reporting period in which it is completed. To build your own interactively and export it, open the Gantt chart creator.

The lump-sum rules add one decision to the Gantt. WP1 and WP6 run the full 42 months, so as drawn they could only be declared complete, and paid, at the final report. The Commission’s lump-sum guidance recommends splitting long work packages along the reporting periods, keeping the same content. For SALTMARSH that turns six work packages into ten: WP1a, WP1b and WP1c for management in RP1, RP2 and RP3, the same for WP6, with WP2 to WP5 unchanged. Each split work package keeps its own row in tables 3.1a and 3.1f and its own lump-sum share. The lump-sum budget example shows the payment schedule that results.

The PERT chart

The PERT answers a different question from the Gantt: what has to finish before what can start. SALTMARSH has a single critical path. The baseline atlas and requirements from WP2 set the design constraints for WP3; the design freeze at MS3 releases WP4 installation; and the installed pilots at MS4 are what WP5 monitors. Any slip on that chain moves the end of the project, which is why the Ravenna permitting risk is rated high severity and why its mitigation starts in WP3, six months before WP4 begins. WP1 and WP6 run alongside, and WP6 carries float until WP5 hands over the model and the policy evidence.

Ten checks before you submit

Templates to copy

WPn  <Title>                                      Lead: <short name>   M<start>–M<end>   <PM> PM
Objectives
  On.1 <What will be true at the end, measurable>
  On.2 ...
Description of work and role of participants
  Tn.1 <Title> (M<a>–M<b>; lead <partner>; <PM> PM)
       <Lead> does <activity, quantified> (<PM> PM). <Partner> does <activity> (<PM> PM).
       Output: <countable result, links to Dn.x>
  Tn.2 ...

MSx  <State of the project that is true or false>   WP<n>   Month <m>   Verified by <named document or artefact>

The tables on this page are plain HTML, so you can copy them straight into a word processor as a starting structure. If you are assembling the work plan from partner contributions written in different formats, that merging step is what Consortium Merge is for. The rest of the consortium logic, from choosing partners to the consortium agreement, is in the consortium survival guide.